Clean heating transitions in China represent a critical socio-technical pathway toward urban decarbonization and air quality improvement. However, heterogeneous economic capacities, resource endowments, and technological readiness have resulted in pronounced regional inequities. This study develops an integrated analytical framework that combines statistical, predictive, and causal approaches to evaluate the progress, effectiveness, and fairness of clean heating implementation between 2017 and 2021. Using exclusive field survey data from renovated households located in the 2 + 26 cities, this study maps the spatiotemporal evolution, assesses reductions in seven key pollutants and constructs a capacity-normalized residual emission burden indicator (heating-related residual emissions per unit of rural income) to characterize regional disparities in transition outcomes. Results reveal that coal-to-gas and coal-to-electricity retrofits accounted for over 90% of conversions, but exhibited diminishing returns in high-intensity regions. Predictive modeling with popular machine learning algorithms and SHAP interpretation indicates that energy structure and intensity emerge as the primary barriers, whereas innovation and fiscal support serve as critical mitigating forces. Finally, Double Machine Learning (DML) method is introduced to estimate conditional average treatment effects (CATE), revealing high heterogeneity in policy outcomes across energy structure levels. These findings underscore the importance of equitable access to clean technology in achieving a just socio-technical transition. The proposed framework offers both theoretical and empirical insights for policymakers and is transferable to other sustainable urban transition contexts.

Publication:

SUSTAINABLE PRODUCTION AND CONSUMPTION

http://dx.doi.org/10.1016/j.spc.2026.05.003

Author:

Li, Linyan

City Univ Hong Kong, Dept Data Sci, Hong Kong 999077, Peoples R China

City Univ Hong Kong, Dept Infect Dis & Publ Hlth, Hong Kong 999077, Peoples R China

Wu, Yunlong

Jiangsu Univ, Sch Environm & Safety Engn, Zhenjiang 212013, Peoples R China

Zhang, Li

Jiangsu Univ, Sch Environm & Safety Engn, Zhenjiang 212013, Peoples R China

Xu, Hui

Jiangsu Univ, Sch Environm & Safety Engn, Zhenjiang 212013, Peoples R China

Wang, Huili(corresponding author)

Chinese Acad Environm Planning, Ctr Carbon Neutral, Beijing 100043, Peoples R China

Email address:wanghl@caep.org.cn

Jin, Ling

Chinese Acad Environm Planning, Ctr Carbon Neutral, Beijing 100043, Peoples R China

Chen, Xiaojun

Chinese Acad Environm Planning, Ctr Carbon Neutral, Beijing 100043, Peoples R China

Lei, Yu

Chinese Acad Environm Planning, Ctr Carbon Neutral, Beijing 100043, Peoples R China

Li, Jieyi

Chinese Acad Sci, Acad Math & Syst Sci, Beijing 100190, Peoples R China

Email address:lijieyi@amss.ac.cn

Xiang, Lin

China Univ Petr, Sch Econ & Management, Qingdao 266580, Peoples R China

Hu, Xiurong

Nanjing Univ Aeronaut & Astronaut, Coll Econ & Management, Nanjing 211106, Peoples R China

He, Jiaxuan

Truman State Univ, Dept Comp Sci & Math, Kirksville, MO USA